The sky doesn't add up
In the last guide you saw the cosmos act as a free particle accelerator, raining cosmic rays on us at energies no machine can match. Now we turn the same telescope on a quieter, stranger fact: when astronomers add up everything they can see — stars, gas, dust, glowing clouds — and ask how much gravity that visible stuff should produce, the answer comes up far too small to explain how the universe actually moves. Something heavy is there that gives off no light at all. This is the dark matter problem, and the word 'dark' is meant literally: it neither shines nor blocks light, it simply pulls.
The simplest clue comes from spinning galaxies. A galaxy is held together by gravity, just like the Solar System: the faster a star orbits, the more mass must be pulling it inward to keep it from flying off. Out at the visible edge of a spiral galaxy, where the stars thin out, we expect orbits to slow down — the way distant Neptune crawls around the Sun far more slowly than Mercury. Instead, the outer stars orbit just as fast as the inner ones. The only honest reading is that the galaxy's gravity does not stop where its light stops. Each galaxy sits inside a vast invisible halo of extra mass, several times heavier than all its stars combined.
Five independent witnesses
What makes the case for dark matter so strong is not one measurement but a stack of them, taken with different instruments, on different scales, that all demand the same missing mass. The gravitational evidence is not a single argument you can wave away; it is several arguments that happen to converge on the same number. Here are the main witnesses, from the size of a galaxy out to the size of the whole universe.
- Galaxy rotation curves. The flat orbital speeds of outer stars, described above. Vera Rubin's careful measurements in the 1970s turned this from a curiosity into a crisis, galaxy after galaxy telling the same story.
- Galaxy clusters. Whole swarms of galaxies orbit one another so fast that their visible mass could never hold the swarm together — it would have flown apart long ago. This was the very first hint, spotted by Fritz Zwicky in the 1930s, who coined the term 'missing mass'.
- Gravitational lensing. Mass bends the path of passing light, so a heavy cluster acts as a giant warped lens, smearing background galaxies into arcs. The amount of bending measures the total mass directly — light doesn't care whether the mass shines — and it again far exceeds the visible matter.
- The Bullet Cluster. Two galaxy clusters that smashed through each other. The hot gas (most of the ordinary matter) piled up in the collision and glows in X-rays, but lensing shows the bulk of the mass sailed straight through, untouched — exactly what you expect if most of the mass is a barely-interacting substance that ignored the crash.
- The cosmic microwave background. The faint relic glow of the infant universe (the subject of a later guide in this rung) carries a pattern of tiny ripples whose sizes encode exactly how much ordinary matter and how much dark matter were present. The fit is precise — and it insists that dark matter outweighs atoms by about five to one.
Could we instead be wrong about gravity itself, rather than missing some matter? It is a fair question, and physicists have tried hard to modify Newton's and Einstein's laws to do without dark matter. Such theories can be tuned to fit single galaxies, but they struggle badly with clusters and especially with the Bullet Cluster, where the mass and the visible gas are plainly in different places. The honest summary: the evidence that something is there is rock solid; what that something is remains wide open.
What could it be made of?
Whatever dark matter is, the evidence already pins down its job description tightly. It must be heavy enough to clump under gravity, yet feel the electromagnetic force so weakly that it neither emits nor absorbs light. It must be stable over the whole age of the universe, or it would have decayed away. And it cannot be ordinary atoms hiding as cold gas or dead stars — the cosmic microwave background and Big Bang nucleosynthesis count the atoms independently, and there simply are not enough. So the leading candidates are new particles, not on the Standard Model's table. This is one of the cleanest pieces of evidence that the Standard Model is incomplete: we know roughly how much of the universe it is missing.
The most famous suspect is the WIMP — a Weakly Interacting Massive Particle. The idea is irresistibly neat. Suppose there is a new particle with a mass somewhere around that of the heavy weak-force carriers (tens to hundreds of GeV) that interacts only through the weak force and gravity. In the hot early universe such particles would have been made and destroyed freely; then, as the universe cooled and thinned, they would have stopped finding each other and 'frozen out' at a leftover abundance set by their interaction strength. When you run that calculation, a weak-strength particle naturally freezes out at almost exactly the dark-matter density we measure. This near-coincidence is called the WIMP miracle, and you will meet the freeze-out machinery itself in the next guide.
The WIMP idea got a second wind from supersymmetry, the proposal that every known particle has a heavier partner. In many such models the lightest of these new partners — the neutralino — is stable, electrically neutral, and weighs in right around the WIMP range, making it a perfect dark-matter candidate almost for free. A second, very different suspect is the axion: an extremely light particle proposed for an unrelated reason, to fix the strong CP problem of why the strong force shows no sign of a certain symmetry violation it is allowed to have. The axion was not invented to be dark matter, but it turns out it could be — and that kind of two-for-one is exactly what makes physicists take an idea seriously.
Three ways to catch a ghost
If dark matter is a real particle that feels the weak force, even a little, then it should occasionally bump into ordinary matter, or annihilate with its own kind, or be created in a collider. Each possibility is the same vertex read in a different direction, and together they define the three great search strategies. This is the direct-versus-indirect detection split, with collider production as a third front.
dark + dark --> ordinary + ordinary read left-to-right (annihilation in space) -> INDIRECT detection read right-to-left (made in a collision) -> COLLIDER production read top-to-bottom (dark hits a nucleus) -> DIRECT detection
Direct detection is the most literal. Our galaxy's dark halo is streaming through your body right now, but so feebly that a given dark particle might pass through the entire Earth without touching a thing. So experimenters build ultra-quiet detectors — tanks of liquid xenon or chilled crystals — deep underground, shielded by a kilometre of rock from cosmic rays, and wait for the rare nudge of a single nucleus recoiling when a dark particle happens to clip it. The signal is a flash of light and a whisper of charge, a few atoms' worth of energy, fished out of a sea of radioactive background. It is a heroic kind of patience.
Indirect detection watches the sky instead. Where dark matter piles up densest — the heart of our galaxy, the centre of the Sun, dwarf galaxies — pairs of dark particles might occasionally meet and annihilate, leaving a tell-tale shower of ordinary particles: gamma rays, antimatter, or neutrinos at a sharp energy. Space telescopes and the kind of ground arrays you met for cosmic rays scan for such excesses. The third front is the collider: at the LHC, two protons might convert their energy into a pair of dark particles, which then sail straight out of the detector unseen. The clue is an imbalance — visible debris flying out one side with nothing to counter it, the missing momentum betraying an invisible passenger that carried energy away.
Decades of silence — and what it means
Here is the honest, uncomfortable heart of the story: all three searches have been running for decades, each has grown enormously more sensitive, and none has found dark matter. Direct-detection experiments have improved their reach by factors of millions, and seen nothing but background. Indirect searches have chased a handful of tantalising excesses — an unexplained gamma-ray glow toward the galactic centre, odd antimatter ratios in cosmic rays — but each, on closer look, has plausible ordinary explanations. The LHC has produced no missing-momentum signal beyond what the Standard Model already predicts from neutrinos. The classic WIMP, in particular, is being squeezed hard: the most natural mass-and-coupling windows are now largely ruled out.
So where does that leave us? With one of the great unsolved problems in physics — the question of what dark matter and dark energy actually are — still genuinely open. The gravitational evidence is among the most secure facts in cosmology, demanding that five-sixths of all matter be invisible. Yet the particle remains uncaught, and the once-favourite WIMP is on the defensive. The response is not despair but breadth: new xenon detectors the size of swimming pools, dedicated axion experiments tuning radios to listen for dark particles converting to photons in a magnetic field, and proposals for entire hidden sectors of particles that talk to us only through the faintest of bridges. The mystery is exactly as honest as it sounds — we know it is out there, we have not yet held it in our hands, and that is precisely why the search goes on.